arXiv:1203.1397v1 [astro-ph.CO] 7 Mar 2012 trfrainhsoyadevrneto h dwarf the of 92 UGCA environment and history formation o.Nt .Ato.Soc. Astron. R. Not. Mon. N Assoc under † the Inc. Space 5-26555. Astronomy, by in NAS the Research contract operated for at Universities is archive of STScI tion data Institute. the Science from Telescope obtained Telescope, Space ⋆ cetdXX eevdXX noiia omXXX form original in XXX; Received XXX. Accepted rusadcod ( clouds and groups infiatpors a enahee ntesuyo re- of study Telescop Space the Hubble ( the in to due achieved populations decad stellar been solved last has the stel In progress study directly. significant to these us of allow population which lar ar , galaxies individual Nearby into galaxies. resolved dwarf studying for convenient motn o nesadn fteroii n evolution. and origin ( Universe their local of The extremel understanding is Universe. galaxies for dwarf the in important in formation star objects of question numerous The most are galaxies Dwarf INTRODUCTION 1 c 3 2 1 4 HST ii Makarova Lidia nvri´ yn1 ilubne -92,Fac;CA,O CRAL, France; Russia Branch, F-69622, SAO Villeurbanne, Chile, 1, Universit´e of Lyon Institute Karac Newton Arkhyz, Isaac Nizhniy Observatory, Astrophysical Special srnmclIsiue an-eesugSaeUniversi State Saint-Petersburg Institute, Astronomical -al [email protected] E-mail: ae nosrain aewt h AAEAHubble NASA/ESA the with made observations on Based X RAS XXX bu 0prcn fnab aaisaestae in situated are galaxies nearby of cent per 50 About n e ag rudbsdtelescopes. ground-based large new and ) aao aahnsv2011 Karachentsev & Makarov 6 0Mc spriual motn and important particularly is Mpc) 10 1 000 , 2 † , mtyMakarov Dmitry , 1 – 11 0 y o30Mrao ti eylkl htteogigsa format star ongoing the that likely very is about It of ago. metallicity Myr higher 300 to Myr 200 otepeet h trfraini hseetsosmdrt e moderate shows event this during in Gyr formed formation 1.5 was star about The mass starting present. formation stellar the star total to recent the of indications of also cent per are 84 About dex. g.Teesasaemsl ea-or ihama ealct [Fe/H metallicity mean a with occurre metal-poor, period mostly activity are r formation stars star formation These main star ago. The the time. construc measured of have We function and as branch. populations galaxy giant stellar The red ga resolved galaxies). old of the of Group including group HST/ACS Local 342 the with (IC of stars extinction vicinity Galactic the in strong dw situated of nearby is the dwarf of irregular history formation This star quantitative a present We ABSTRACT tla otn aais niiul GA92 UGCA individual: galaxies: – content stellar n th dwarf words: history. closest suggest Key formation galaxy’s We star the galaxies. their to affect related two not not con these is does temporal 1569 in or NGC causal events in no phenomenon f reveals formation star 1569 NGC star our of Comparing recent that 1569. with NGC 92 UGCA galaxy of starburst the to companion XX rne 2Nvme 08(NL (MN 2018 November 12 Printed (XXX) ⋆ aais wr aais omto aais vlto galaxies: – evolution galaxies: – formation galaxies: – dwarf galaxies: .Taking ). y an-eesug Russia Saint-Petersburg, ty, ASA a-hresa396,Russia 369167, hai-Cherkessia es, ia- 1 y e e , - sraor eLo,S.GnsLvl -96,France F-69561, Laval, Genis St. Lyon, de bservatoire 2 , 3 egySavchenko Sergey , − 0 of 1994b ( galaxies dwarf 2006 of associations loose account into Nilson of neighbour closest ( the 1569 NGC be for galaxy to has starburst considered which nearest the 92, been UGCA time galaxy long irregular his- a formation dwarf star the In the of considered complex. tory have 342/Maffei we IC work within present the galaxies dwarf of images ( Universe nearby challenge. a main is of it determination behind a galaxies and Avoid- of us of from properties evolution region ‘Zone this the and hides Unfortunately, ance’ dynamic Universe. nearby on the impact of significant have should lntr euaby planetary strong by Fig. obscured (see is extinction 342 IC Galactic around group nearby The jects. . – 6 ose tal. et Hoessel ntefaeoko h td ftesrcueo the of structure the of study the of framework the In h ml reua aayUC 2wsdsoee by discovered was 92 UGCA galaxy irregular small The ,ms fteglxe ihn3Mcaentioae ob- isolated not are Mpc 3 within galaxies the of most ), − ; aaoa&Krcete 2003 Karachentsev & Makarova ( 0 1974 . e.UC 2i fe osdrdt ethe be to considered often is 92 UGCA dex. 3 n neednl aaouda possible a as catalogued independently and ) A T E ( HST tl l v2.2) file style X 1988 li tal. et Ellis atal eovn tit individ- into it resolving partially ) rjc 71 ehv bandthe obtained have we 9771) project 4 1 .TeI 4-ae complex 342-Maffei IC The ). ( 1984 r aayUC 92. UGCA galaxy arf hneetfrom nhancement .CDobservations CCD ). ). g n continuing and ago t n metallicity and ate bu 4Gyr 14 – 8 about d aahnsve al. et Karachentsev hseet There event. this a eovdinto resolved was rainhistory ormation ttestarburst the at eto between nection ] aisi zone a in laxies e oe of model a ted o eidhas period ion − ∼ ihor and eighbours 1 ul tal. et Tully . – 5 − 2 ∼ . 0 2 Makarova et al.

Figure 1. Location of the IC 342/Maffei complex of galaxies in Galactic coordinates. Colours denote galaxy type. The IRAS extinction map is shown in grey scale.

ual stars in the g and r passband images first showed that Table 1. General parameters of UGCA 92. UGCA 92 (EGB 0427+63) is a dwarf irregular galaxy. Hodge h m s & Miller (1995) detected 25 H ii regions within UGCA 92, R.A.(J2000) 04 32 03.5 [2] Dec (J2000) +63◦36′58′′ [2] concentrated in two well-separated regions of the galaxy. The reddening, measured from the emission-line spectra for size, arcmin 2.0 × 1.0 [3] bright H ii regions is E(B − V ) = 0.90 ± 0.08 and the mean Linear diameter, kpc 3.1 [1] oxygen abundance is about 13 per cent solar, with an uncer- (m − M)0, mag 27.41 ± 0.25 [1] tainty of 50 per cent. Karachentsev & Kaisin (2010) carried Distance, Mpc 3.03 ± 0.35 [1] out Hα flux measurement for UGCA 92 and derived a cur- BT , mag 15.22 [4] rent star formation rate of log(SFR) = −1.51 M⊙/yr. (B − V )T , mag 1.34 [4] The general parameters of UGCA 92 are presented in V3′ , mag 14.55 [5] Table 1. The total magnitudes, colours and central surface I3′ , mag 12.87 [5] JT , mag 12.38 [6] brightnesses, µ(0), are not corrected for Galactic extinction. KsT , mag 11.12 [6] −2 µ(0)B , mag arcsec 25.1 [4] −2 µ(0)V , mag arcsec 24.18 ± 0.01 [5] −2 2 THE DATASET µ(0)I , mag arcsec 22.61 ± 0.01 [5] E(B − V ), mag 0.79 ± 0.13 [7] The dwarf irregular galaxy UGCA 92 was observed aboard AI , mag 1.54 ± 0.25 [7] HST using Advanced Camera for Surveys (ACS) at March −1 28, 2004 (SNAP 9771, PI I. Karachentsev). Two exposures VLG, kms 93 [8] MB, mag −15.61 [1] were made with the filters F606W (1200 s) and F814W (900 8 MHI ,M⊙ 1.56 × 10 [8] s). An ACS image of the dwarf galaxy is shown in Fig. 2. MHI /LB 0.55 [8] The photometry of resolved stars in the galaxy was per- 1 Fraction of old stars (12–14 Gyr), % 84 ± 7 [1] formed with the ACS module of the dolphot package for Metallicity of old stars, [Fe/H], dex −2.0 – −1.5 [1] crowded field photometry (Dolphin 2002) using the recom- Fraction of young stars (500 Myr), % 7.6 ± 0.7 [1] −1 −1 mended recipe and parameters. Only stars with photometry hSFRi, 12 – 14 Gyr ago, M⊙ yr 1.2 ± 0.1 × 10 [1] −1 −2 of good quality were included in the final list, following rec- hSFRi, last 500 Myr, M⊙ yr 4.3 ± 0.6 × 10 [1] ommendations given in the dolphot User’s Guide. We have selected stars with signal-to-noise (S/N) > 5 in both filters, [1] this work; [2] LEDA; [3] Karachentsev et al. (2004); 2 [4] Karachentseva et al. (1996); [5] Sharina et al. (2008); χ 6 2.5 and |sharp| 6 0.3. The resulting colour-magnitude [6] Vaduvescu et al. (2005); [7] Schlegel et al. (1998); [8] Be- diagram contains 32699 stars (Fig. 3). gum et al. (2008) Artificial star tests were performed using the same re- duction procedures to estimate photometric errors, crowd- edness and blending effects in the most accurate way. A large library of artificial stars was generated spanning the necessary range of stellar magnitudes and colours so that the distribution of the recovered photometry is adequately 1 http://purcell.as.arizona.edu/dolphot/ sampled. The photometric errors and completeness are pre-

c XXX RAS, MNRAS 000, 1–11 Star formation history of UGCA92 3

Figure 2. HST /ACS image of UGCA 92 in F606W filter. The image size is 3.4 × 3.4 arcmin.

19 20 21 22 23 24 25 26 27 28 29 16 18 20 22 24 26 28 1 1

0.8 0.8

0.6 0.6

0.4 0.4

completeness 0.2 completeness 0.2

0 0

−1 −0.5

−0.5 , mag , mag

∆ ∆ 0 0

0.5 0.5 19 20 21 22 23 24 25 26 27 28 29 16 18 20 22 24 26 28 F606W, mag F814W, mag

Figure 4. Photometric errors and completeness for UGCA 92. The top panels show the completeness, i.e. the fraction of artificial stars recovered within the photometric reduction procedure, as a function of the F606W and F814W magnitudes. The bottom panels give the difference between the measured and the true input magnitude (∆mag = measure − input). The error bars are 1 σ residuals. sented in Fig. 4. The 1 σ photometric precision is 0.07 at 3 COLOUR-MAGNITUDE DIAGRAM F814W = 25 and 0.19 at F814W = 27 mag. Malmquist bias becomes notable for stars with F814W > 26.8 mag. In All resolved stars are significantly shifted to redder colours F606W the 1 σ photometric precision is 0.06 at 26 and 0.18 due to high extinction in the Zone of Avoidance of the at 28 mag. Milky Way. The diagram is typical for dwarf irregular galax- ies. A pronounced upper main sequence (MS) and proba- ble helium-burning blue loop stars are found at F606W −

c XXX RAS, MNRAS 000, 1–11 4 Makarova et al.

|(V −I)0| 6 0.5 and 19.5 6 I0 6 24.0. According to our mea- surements, the mean MS colour is (V −I)0 = 0.04 mag. The colour spread of the MS ∆(V − I)0 = 0.20 mag is rather 19 large. This spread, as well as the slightly reddish mean MS colour, could indicate a presence of a blue loop population. Thereby, the extinction value given by Schlegel et al. (1998) 20 seems reasonable, and we also do not expect a significant value of internal extinction in this faint dwarf galaxy. The reddening in UGCA 92 was also determined by 21 Hodge & Miller (1995) from the spectroscopy of 25 bright H ii regions within the galaxy. Their reddening value E(B − V ) = 0.90 ± 0.08 is in agreement with the value of Schlegel 22 et al. (1998) within uncertainties. The resolution in the 3 IRAS/DIRBE maps is 6.1 arcmin (Schlegel et al. 1998). It mag 1 is considerably larger than the ACS field of view. Therefore, we cannot get an information about possible variations in 23 2

F814W, extinction across the field of view from these maps. How- ever, a simple test was made using a colour-magnitude dia- 24 gram of UGCA 92. A value of the tip of the red giant branch was measured in four different subframes along y axis of the whole ACS frame. We have found, that the TRGB value has no variations within 1.5 σ of the TRGB uncertainty. Con- 25 sequently, we suggest no variations of external extinction within the ACS field of UGCA 92. Therefore, we use the colour excess E(B − V ) = 0.79 ± 26 0.13 from Schlegel et al. (1998) in all our measurements in the present paper. 4 27 0 1 2 3 3.2 Foreground contamination F606W−F814W, mag The colour-magnitude diagram is highly contaminated by Figure 3. The (F606W − F814W), F814W CMD of the dwarf Milky Way (MW) stars. To account for this contamination galaxy UGCA 92. In the dense parts of the diagram the colour we need to construct colour-magnitude diagram of the MW represents the density in the Hess diagram, while individual in the direction of UGCA 92. This information can be ob- stars are represented where they can be individually distin- tained from images of neighbouring ‘empty’ fields. Three guished. The magnitudes are not corrected for Galactic extinc- fields nearby UGCA 92 were found in the HST data archive tion. Padova isochrones (Girardi et al. 2000) corresponding to exposed with WFPC2 in the parallel mode (coordinates of the mean age and metallicity of detected star formation episodes the centre are 4h29m53s, +64◦42′34′′). However, the pho- are shown: ‘1’ is Z=0.008,t=10 Myr; ‘2’ – Z=0.001,t=50 Myr; ‘3’ tometric limit of these images is about 3 mag higher than – Z=0.0004,t=150 Myr; ‘4’ – Z=0.0004,t=13 Gyr. photometric limit of the working UGCA 92 images. To ac- count for the MW contamination in the region of faint stars, we have used the trilegal program (Girardi et al. 2005), F814W < 1.2 mag. The red supergiant branch (RSG) and which computes synthetic colour-magnitude diagrams for the intermediate age asymptotic giant branch (AGB) are the specified coordinates in the sky and given parameters also well populated. The most abundant feature in the CMD of the Milky Way models. Consequently, we use neighbour- is the red giant branch (RGB) (see Fig. 3). ing field stars to account for the Milky Way contribution to the UGCA 92 CMD brighter than F814W = 24 mag and the trilegal 3.1 Extinction synthetic data for the fainter part of the CMD. Thirty synthetic CMDs were constructed with trilegal The dwarf galaxy UGCA 92 is situated at the Galactic lati- and then averaged to avoid stochastic errors in synthetic ◦ tude l = +10.5 . Galactic extinction in the field of UGCA 92 CMDs. Random and systematic photometric uncertainties needs to be addressed in more detail because reddening es- and completeness measured from artificial star experiments timation at low Galactic latitude in the Zone of Avoidance were applied to the synthetic CMDs. Resulting contamina- is highly uncertain (see Schlegel et al. 1998, appendix C). tion by the MW stars was determined to be 202 stars over Schlegel et al. (1998) give a colour excess of E(B − V ) = the CMD of UGCA 92. 0.79 ± 0.13 using IRAS/DIRBE maps of infrared dust emis- sion. We apply this value to the colour-magnitude diagram and estimate a mean colour of the upper MS as a first test 4 DISTANCE DETERMINATION of the feasibility of the colour excess value. The unreddened V − I colour of upper MS stars is about zero (see Kenyon The distance of UGCA 92 was first estimated by Karachent- & Hartmann 1995, for example). We have selected main se- sev et al. (1994a) using the brightest blue and red super- quence stars in the appropriate magnitude and colour range, giants as distance indicators. The estimated distance mod-

c XXX RAS, MNRAS 000, 1–11 Star formation history of UGCA92 5

18 20 22 24 26 28 18 1

UGCA 92 19 0.5

20 0 −1.5

−1 21 −0.5

0 22 0.5 18 20 22 24 26 28 23 F814W (mag)

3 10 F814W (mag) 24

24.84 25

2 10 26

27

1 28 10 −1 0 1 2 3 24 24.2 24.4 24.6 24.8 25 25.2 25.4 25.6 25.8 F606W−F814W (mag) F814W (mag)

Figure 5. The colour-magnitude diagram (left panel) and the TRGB calculation results for UGCA 92. The upper right panel shows the completeness, photometric errors and dispersion in errors (vertical bars) versus the HST /ACS F814W filter magnitude. The lower right panel gives a histogram of the F814W luminosity function. The resulting model LF convolved with photometric errors and incompleteness is displayed as a bold solid line with a jump at the position of the TRGB.

ulus (m − M)0 = 26.72 mag suffered from large photomet- 5 STAR FORMATION HISTORY ric uncertainties associated with crowded stellar fields and uncertain reddening. Later Karachentsev et al. (1997) esti- The star-formation and metal-enrichment history of mated a distance to UGCA 92 using the same distance indi- UGCA 92 has been determined from its CMD using our cators and better quality data. The resulting distance modu- StarProbe package. This program adjusts the observed lus is (m−M)0 = 26.25 mag. Recently, more precise distance photometric distribution of stars in the colour-magnitude determination was made using HST /ACS data (Karachent- diagram against a positive linear combination of synthetic sev et al. 2006). A deep colour-magnitude diagram permit- diagrams of single stellar populations (SSPs, single age and ted the use of the TRGB distance indicator, resulting in a single metallicity). Our approach is described in more detail distance modulus of (m − M)0 = 27.39 mag. in Makarov & Makarova (2004) and Makarova et al. (2010). However, the TRGB method has recently been con- The observed data were binned into Hess diagrams, giv- siderably improved. We have determined the photometric ing the number of stars in cells of the CMD (two-dimensional TRGB distance with our trgbtool program, which uses histogram). The size of the cell is 0.05 mag in each passband. a maximum-likelihood algorithm to determine the magni- The synthetic Hess diagrams were constructed from theoret- tude of the tip of the red giant branch from the stellar ical stellar isochrones and initial mass function (IMF). Each luminosity function (Makarov et al. 2006). The estimated artificial diagram is a map of probabilities to find a star in a value of TRGB is equal to F814W = 24.84 ± 0.02 mag cell for given age and metallicity. We used the Padova2000 in the ACS instrumental system. The calibration of the set of theoretical isochrones (Girardi et al. 2000), and a TRGB distance indicator has also recently been improved Salpeter (1955) IMF. The distance is adopted from the (Rizzi et al. 2007), where the colour dependence of the ab- present paper (see above) and the Galactic extinction from solute magnitude of the TRGB and zero-point issues in Schlegel et al. (1998). The synthetic diagrams were altered HST /ACS and WFPC2 have been addressed. Using this by the same incompleteness and crowding effects, and the calibration, we have obtained the true distance modulus to photometric systematics, as those determined for the obser- UGCA 92 of (m − M)0 = 27.41 ± 0.25 mag and a distance of vations using artificial stars experiments. We also have taken D = 3.03±0.35 Mpc. Bear in mind that the given small error into account the presence of unresolved binary stars (binary of the TRGB measurement (0.02 mag) and the high preci- fraction). Following Barmina et al. (2002), the binary frac- sion of the calibration (0.02 mag), the resulting accuracy tion was taken to be 30 per cent. The mass distribution for is entirely determined by uncertainty of foreground extinc- the secondary was taken to be flat in the range 0.7 to 1.0 tion of 0.25 mag in the direction of UGCA 92. The colour- of the primary mass. The best-fitting combination of syn- magnitude diagram with the fitted RGB luminosity function thetic CMDs is a maximum-likelihood solution taking into and the resulting TRGB value is shown in Fig. 5. account the Poisson noise of star counts in the cells of the

c XXX RAS, MNRAS 000, 1–11 6 Makarova et al.

Hess diagram. The resulting star formation history (SFH) ages in range ∼10 Myr – 1 Gyr, (lower left panel) was quite is shown in the Fig. 6. The 1 σ error of each SSP is derived conservative (1.0 6 (V − I)0 6 1.5 and I0 6 21.5) to avoid from an analysis of likelihood function. possible contamination of the spatial structure by young and According to our measurements, the main star forma- intermediate age AGB stars (which are redder) and RGB tion event occurred in the period 12 – 14 Gyr ago with a (which are fainter). As a result, the selected population is rather high mean star formation rate (SFR) of 1.2 ± 0.1 × not numerous and smoothly distributed in the central part −1 −1 10 M⊙ yr . This is the total SFR over the whole galaxy. of the galaxy nearly along the major axis. The metallicity range for the most stars formed during this A most numerous population is the red giant branch event is [Fe/H] = [−2.0 : −1.5] dex. This initial burst ac- (>1 Gyr) observed nearly (V − I)0 > 0.5 and I0 > 23.0. counts for about 84±7 per cent of the total mass of formed The stars are widely distributed in the image with apparent stars. concentration to the centre of UGCA 92 and nearly the ma- The quiescence period has appearing from about 6 to jor axis of the object. The density of the stars is smoothly 12 Gyr ago. However, the absence of star formation activity decayed to the edge of the galaxy, which probably extends in this period could be due to tight packing of the different beyond the image boundary. age stars in the upper part of the RGB. At the distance of 3 Mpc, fainter stellar populations, like a horizontal branch and a lower part of the main sequence, are hardly resolved at the 6 STAR FORMATION AND ENVIRONMENT HST. Without these details it is difficult to resolve an age- metallicity-SFR relation for the oldest ( > 6 ÷ 8 Gyr) star Maffei – IC 342 is highly obscured nearby galaxy complex. formation events, due to tight packing of the correspondent These galaxies are gathered in two groups. One group is isochrones for the brightest part of the CMD. around the giant face-on galaxy IC 342 and another one is There are signs of marginal (insignificant) star forma- around the pair of E+S galaxies Maffei 1 and Maffei 2. Ac- tion 4 – 6 Gyr ago. A metallicity of these stars is similar to cording to Karachentsev (2005), the groups contain eight the metal abundance of the oldest stellar population. members each. One should keep in mind, that the distances There are also indications of recent star formation start- to the group members could be highly uncertain due to the ing about 1.5 – 2 Gyr ago and continuing to the present. We high uncertainty in the Galactic extinction in this direction. have measured the SFH in short age periods in the last 500 The IC 342 group structure is shown in the Fig. 8. There Myr to give more detail for the recent and ongoing star for- are 19 galaxies in the plot which situated at distances 6 1.8 mation. A mean SFR of the stars formed in the last 50 Myr Mpc from IC 342. The distances were taken from the Cata- −2 −1 is 3.9 ± 0.6 × 10 M⊙ yr . This ongoing star formation log of Neighboring Galaxies (Karachentsev et al. 2004). The rate is in good agreement with the independent estimation data on the distances were updated recently (Karachentsev, by Karachentsev & Kaisin (2010) from Hα flux measure- private communications). The two galaxy groups, around −2 −1 ments (3.2 ± 0.3 × 10 M⊙ yr ). The recent star forma- IC 342 and around Maffei 1 and Maffei 2 could be seen in tion is showing moderate enhancement from ∼ 200 Myr to the figure. Most of the complex members are dwarf galaxies. 300 Myr. A mean star formation rate in the last 500 Myr is The morphological types of the objects are coded by a colour −2 −1 4.3 ± 0.6 × 10 M⊙ yr . A mass portion of stars formed in according to de Vaucouleurs et al. (1991) and Karachentsev the last 500 Myr is 7.6±0.7 per cent of the total stellar mass. et al. (2004) from the early types (red) to the late types A metallicity of the recent star formation event is determin- (dark blue). It is interesting to note that all the dwarf satel- ing with large uncertainty due to relatively poor statistic in lites of IC 342, including the galaxy under study UGCA 92 comparison with sufficiently more numerous old stars. How- are irregulars. An absence of dwarf spheroidal satellites sub- ever, the measurements show that a significant part of the system in the group can imply that dSphs still were not young stars is evidently metal enriched. It is very likely that discovered, because their lower surface brightness and high the ongoing star formation has a metallicity of −0.6 – −0.3 Galactic extinction put serious observational constrains. dex. The dwarf irregular galaxy under study UGCA 92 is situated at a linear distance D = 440 kpc from the grav- itational centre of the group IC 342. The three nearest to 5.1 Spatial distribution of stellar populations UGCA 92 galaxies are UGCA 86 (a linear distance D = 260 We have selected few stellar populations depending on their kpc), UGCA 105 (D = 300 kpc) and NGC 1569 (D = 360 position in the CM diagram. The result of this selection is kpc). UGCA 92 is often considered to be the companion to presented in Fig. 7. The upper main sequence stars (upper the starburst galaxy NGC 1569. They have been known to −1 left panel) has the unreddened colour (V − I)0 6 0, their have close radial velocities, VLG = 93 km s for UGCA 92 −1 age range is about 10–100 Myr. These youngest stars of the (Begum et al. 2008) and VLG = 106 km s for NGC 1569 galaxy are mainly concentrated in few knots which highlight (Walter et al. 2008). Indeed, except for the giant spiral regions of ongoing star formation and form actual irregular IC 342 (D = 320 kpc from NGC 1569) and a very small structure of UGCA 92. irregular galaxy Cam B (D = 190 kpc from NGC 1569), In the upper right panel of the figure we show supposed UGCA 92 is the spatially closest companion to NGC 1569. blue-loop (BL) stars selected by a colour 0.0 6 (V − I)0 6 A median radial velocity of galaxies within IC 342 group −1 0.5. Their age range is ∼10–200 Myr. Spatial distribution is VLG = 244 km s with the velocity dispersion of 79 of these stars is considerably more smooth and regular in km s−1. The pair of galaxies NGC 1569 – UGCA 92 has a comparison with MS stars though they are well-concentrated maximal peculiar velocity within the group. The question to the central parts of the galaxy. arises whether these galaxies form a real subsystem within A selection criterion for red supergiant stars (RS), with IC 342 group. Indeed, the uncertainty in UGCA 92 distance

c XXX RAS, MNRAS 000, 1–11 Star formation history of UGCA92 7

8 9 10 10 10 10 0.15

0.1 −2.5 −2 −1.5 −1 log( SFR ) SFR 0.05

0

−0.4 −0.4

−0.6 −0.6

−0.8 −0.8

−1 −1

[Fe/H] −1.2 −1.2

−1.4 −1.4

−1.6 −1.6

−1.8 −1.8

8 9 10 10 10 10 0 0.05 0.1 Age, yr

Figure 6. The star formation history of the dwarf irregular galaxy UGCA 92. Top panel shows the star formation rate (SFR) (M⊙/yr) against the age of the stellar populations. The bottom panel represents the metallicity of stellar content as function of age. The colour corresponds to the strength of SFR for given age and metallicity. is 350 kpc (see Tab. 1), which is pretty close to derived lin- no evident age/metallicity gradient, i.e. the starburst phe- ear separation between NGC 1569 and UGCA 92. The main nomenon is highly centrally concentrated. Probably, such a source of this uncertainty is a huge Galactic absorption. Tak- morphology could be the result of past interaction/merging. ing into account the close radial velocities and close positions The question arises whether we can find any signs of past on the sky, these galaxies could form a tight physical pair. interactions for the dwarf galaxies using the information on In this case we could expect to find a correlation in their star formation of these objects. star formation. From the other hand, the estimation of the Besides UGCA 92, in the subgroup of the clos- linear distance between the galaxies is reasonable. There- est neighbours NGC 1569–UGCA 92–UGCA 86–UGCA 105 fore, the similar velocities and positions on the sky are just only UGCA 86 has observations deep enough (HST /ACS coincidence in a virialized system. images within the project number 9771, PI I. Karachent- sev) to judge about overall star formation from the pho- NGC 1569 is the well-known nearest strong starburst tometry of resolved stars. The colour-magnitude diagram of galaxy. Vallenari & Bomans (1996) have found evidence of a this galaxy is presented in the Fig. 9. The both UGCA 92 recent burst of star formation from about 15 to 4 Myr ago. (D = 3.03 Mpc) and UGCA 86 (D = 2.96 Mpc, Karachent- One more distinct burst was found not older than 150 Myr sev et al. (2006)) are at nearly the same distance from us. ago. Later Angeretti et al. (2005) have found three recent However, UGCA 86 has the brighter absolute stellar mag- distinct starbursts using HST/NICMOS data: 13 – 37 Myr, nitude (MB = −17.95) and larger angular size. We have 40 – 150 Myr and ∼ 1 Gyr ago (for 2.2 Mpc distance). A measured more resolved stars in the ACS field. An up- presence of bright H ii regions also indicate substantial on- per main sequence and helium-burning blue loop stars are going star formation. The galaxy contains two extremely found at F606W − F814W < 1.6 mag. The well-populated luminous super-star clusters (Hunter et al. 2000). In the RSG and the huge AGB are situated above the TRGB at last work 45 other young clusters also have been identified, F606W − F814W > 1.6 mag and F814W < 25.14 mag. the most of them have an age < 30 Myr. The distance to Galactic extinction for UGCA 86 is even higher (E(B−V )= this galaxy was uncertain for the long time. The accurate 0.94 mag according to Schlegel et al. (1998)). A photometric TRGB distance (3.36±0.20 Mpc) was measured by Grochol- limit and the large colour excess seriously affect the RGB ski et al. (2008) using HST /ACS images. Stellar content of zone in the CMD (see Fig. 9). This fact reduces seriously a the NGC 1569 halo was studied in great detail recently by reliability of a computational modelling of the star forma- Ry´set al. (2011). Judging from these two papers, the old tion older than 1 Gyr. Therefore, we only fitted a number (about 10 Gyr) RGB stars, including the outer halo in this of theoretical isochrones to the young stellar population of starburst galaxy should be more metal reach ([Fe/H] ≃−1) UGCA 86. The age and metallicity of these populations are than we measured for the old RGB stars of UGCA 92. One similar in UGCA 86 and UGCA 92. of the interesting conclusions of Ry´set al. (2011) is that the outer stellar halo of the starburst galaxy NGC 1569 is H i and Hα observations of NGC 1569, UGCA 92 and not tidally truncated and it is not outward extension of the UGCA 86 were performed by different authors. Numerous inner disk, but instead it is the distinct stellar halo with Hα knots were detected in all the three galaxies, tracking

c XXX RAS, MNRAS 000, 1–11 8 Makarova et al.

Figure 7. Spatial distribution of the stellar populations in UGCA 92. The main sequence stars is bounded by (V − I)0 6 0. The boundaries of the shown blue loop are 0.0 6 (V − I)0 6 0.5. RS stars correspond to the range 1.0 > (V − I)0 6 1.5 and I0 6 21.5. The red giant population bounded by (V − I)0 > 0.5 and I0 > 23.0. Stellar density contours of the populations are shown with the solid lines. the ongoing star formation events in these objects (Hodge of UGCA 86, with two separate components: a rotating disk & Miller 1995; Kingsburgh & McCall 1998; Karachentsev & and a highly elongated spur that is kinematically disjunct Kaisin 2010). High-sensitivity H i maps of NGC 1569 show from the disk. i an evidence for a companion H cloud connected with the Hereby, summarizing the cited results, we could con- i galaxy by a low surface brightness H bridge. At the edge clude, that hydrogen at the outskirts of the starburst galaxy i of NGC 1569 it coincides with H arcs (Stil & Israel 1998). NGC 1569 is highly disturbed with the signs of infall, but no The hydrogen cloud is apparently not correlated with any similar features were detected in the neighbouring dwarfs optical satellites/counterparts. Stil & Israel (2002) argued, UGCA 92 and UGCA 86. that about 10 per cent by mass of all H i in NGC 1569 have unusually high velocities. Some of this H i may be associ- It should be note, that all the data known to date could ated with the mass outflow evident from Hα measurements, not give us the particular age, when the series of the distinct but some may also be associated with the NGC 1569’s H i intensive star bursts in NGC 1569 was started. Angeretti companion and H i bridge, in which case, infall rather than et al. (2005) mention that the last starburst should occur 8 outflow might be the cause of the discrepant velocities. H i – 27 Myr ago if the distance to the galaxy is 2.9 Mpc instead observations by Stil et al. (2005) show a complex structure 13 – 37 Myr assuming the distance of 2.2 Mpc. The revised accurate distance to NGC 1569 is 3.36 Mpc according to

c XXX RAS, MNRAS 000, 1–11 Star formation history of UGCA92 9

Figure 8. The 3D structure of the IC 342 group. The size of the data cube is 1.8 × 1.8 × 1.8 Mpc. The giant spiral IC 342 is placed in the centre of the cube.

Grocholski et al. (2008). It is possible that other two star- are also signs of marginal star formation 4 – 6 Gyr ago. A burst periods occurred 40 – 150 Myr ago and about 1 Gyr metallicity of these stars is similar to the metal abundance ago will be shifted to the younger ages assuming the new dis- of the oldest stellar population. tance. Our data on recent star formation in the companion UGCA 92 galaxy show substantial and continuous star for- UGCA 92 has a typical morphology of an irregular mation in the last 500 Myr with the enhancement at about dwarf with apparent associations of bright blue stars in its 200 – 300 Myr. Therefore, our results do not indicate a di- body. Numerous H ii knots were also detected in the galaxy, rect connection between the recent star formation in the two tracking the ongoing star formation. According to our mea- galaxies. surements, recent star formation was started about 1.5 – 2 Gyr ago and continuing to the present. We modelled the star formation history with good time resolution for the re- 7 CONCLUSIONS cent star formation event. The continuous star formation in this period shows moderate enhancement from about 200 We have derived a quantitative star formation history of Myr to 300 Myr ago. A mean star formation rate in the last −2 −1 −2 the dwarf irregular galaxy UGCA 92 situated in the highly 500 Myr is 4.3 ± 0.6 × 10 M⊙ yr and 3.9 ± 0.6 × 10 −1 obscured nearby galaxy group IC 342. Due to a low Galactic M⊙ yr in the last 50 Myr. The mass portion of the stars latitude (l = +10.5◦) the extinction is very high (E(B − formed in the last 500 Myr is 7.6 per cent of the total mass V ) = 0.79 ± 0.13 according to Schlegel et al. (1998)) in of formed stars. A metallicity of the recent star formation this direction. The star formation history was reconstructed event is determining with a large uncertainty due to rel- using HST /ACS images of the galaxy and a resolved stellar atively poor statistic in comparison with sufficiently more population modelling. According to our measurements, 84 numerous old stars. However, the measurements show that per cent of the total stellar mass were formed during the star a significant part of the young stars is evidently metal en- formation occurred about 12 – 14 Gyr ago. A metallicity riched. It is very likely that the ongoing star formation has range of these stars is [Fe/H] = [−2.0 : −1.5] dex. There a metallicity of −0.6 – −0.3 dex.

c XXX RAS, MNRAS 000, 1–11 10 Makarova et al.

18 ACKNOWLEDGEMENTS The work was supported by the Russian Foundation for Basic Research (RFBR) grant 11–02–00639 and Russian- 19 Ukrainian RFBR grant 11–02–90449. We acknowledge the support of the Ministry of Education and Science of the Rus- sian Federation, the contract 14.740.11.0901. This work was 20 partially supported by Physical Sciences Department pro- gram of Russian Academy of Sciences. We are thankful to the anonymous referee for the very useful comments to the 21 paper and to Scott Trager for his kind help with the text preparation. We acknowledge the usage of the HyperLEDA database (http://leda.univ-lyon1.fr). 22

mag 3 1 23 REFERENCES

F814W, 2 Angeretti L., Tosi M., Greggio L., Sabbi E., Aloisi A., Lei- therer C., 2005, AJ, 129, 2203 24 Barmina R., Girardi L., Chiosi C., 2002, A&A, 385, 847 Begum A., Chengalur J. N., Karachentsev I. D., Sharina M. E., Kaisin S. S., 2008, MNRAS, 386, 1667 25 de Vaucouleurs G., de Vaucouleurs A., Corwin Jr. H. G., Buta R. J., Paturel G., Fouque P., 1991, Third Reference Catalogue of Bright Galaxies, de Vaucouleurs, G., de Vau- 26 couleurs, A., Corwin, H. G., Jr., Buta, R. J., Paturel, G., & Fouque, P., ed. 4 Dolphin A. E., 2002, MNRAS, 332, 91 27 Ellis G. L., Grayson E. T., Bond H. E., 1984, PASP, 96, 0 1 2 3 283 F606W−F814W, mag Girardi L., Bressan A., Bertelli G., Chiosi C., 2000, A&AS, 141, 371 Figure 9. F606W F814W F814W The ( − ), CMD of the dwarf Girardi L., Groenewegen M. A. T., Hatziminaoglou E., da galaxy UGCA 86. The magnitudes are not corrected for Galactic Costa L., 2005, A&A, 436, 895 extinction. The same Padova theoretical isochrones (Girardi et al. 2000) as for UGCA 92 CMD were shown. Grocholski A. J., Aloisi A., van der Marel R. P., Mack J., Annibali F., Angeretti L., Greggio L., Held E. V., Romano D., Sirianni M., Tosi M., 2008, ApJ, 686, L79 UGCA 92 is often considered to be the companion to Hodge P., Miller B. W., 1995, ApJ, 451, 176 the starburst galaxy NGC 1569. They have been known to Hoessel J. G., Saha A., Danielson G. E., 1988, PASP, 100, −1 have close radial velocities, VLG = 93 km s for UGCA 92 680 −1 (Begum et al. 2008) and VLG = 106 km s for NGC 1569 Hunter D. A., O’Connell R. W., Gallagher J. S., Smecker- (Walter et al. 2008). The linear distance between the galax- Hane T. A., 2000, AJ, 120, 2383 ies is D = 360 kpc. These two objects are evidently could Karachentsev I., Drozdovsky I., Kajsin S., Takalo L. O., be considered as a close pair of galaxies. Heinamaki P., Valtonen M., 1997, A&AS, 124, 559 Another dwarf galaxy close to UGCA 92 is UGCA 86 Karachentsev I. D., 2005, AJ, 129, 178 with the linear distance D = 260 kpc. Our HST/ACS data Karachentsev I. D., Dolphin A., Tully R. B., Sharina M., allow to judge about overall star formation from the photom- Makarova L., Makarov D., Karachentseva V., Sakai S., etry of resolved stars. Theoretical isochrone fitting shows an Shaya E. J., 2006, AJ, 131, 1361 apparent similarity of the resolved stellar populations in the Karachentsev I. D., Kaisin S. S., 2010, AJ, 140, 1241 two galaxies. Karachentsev I. D., Karachentseva V. E., Huchtmeier It is worth to note, that the mean metallicity of the old W. K., Makarov D. I., 2004, AJ, 127, 2031 RGB stars measured by us in UGCA 92 is lower, than the Karachentsev I. D., Tikhonov N. A., Sazonova L. N., 1994a, known metallicity of the halo RGB stars in NGC 1569. Astronomy Letters, 20, 84 Comparing our star formation history of UGCA 92 with —, 1994b, A&AS, 106, 555 that of NGC 1569 reveals no causal or temporal connection Karachentseva V. E., Prugniel P., Vennik J., Richter G. M., between recent star formation events in these two galaxies. Thuan T. X., Martin J. M., 1996, A&AS, 117, 343 We suggest, that the starburst phenomenon in NGC 1569 Kenyon S. J., Hartmann L., 1995, ApJS, 101, 117 has not related to the closest dwarf neighbours and does Kingsburgh R. L., McCall M. L., 1998, AJ, 116, 2246 not affect their star formation history. Makarov D., Karachentsev I., 2011, MNRAS, 412, 2498 Probably, detailed N-body modelling of the group Makarov D., Makarova L., Rizzi L., Tully R. B., Dolphin within 300 kpc from IC 342 is necessary to clarify a reason A. E., Sakai S., Shaya E. J., 2006, AJ, 132, 2729 of recent star bursts in NGC 1569. Makarov D. I., Makarova L. N., 2004, Astrophysics, 47, 229

c XXX RAS, MNRAS 000, 1–11 Star formation history of UGCA92 11

Makarova L., Koleva M., Makarov D., Prugniel P., 2010, MNRAS, 406, 1152 Makarova L. N., Karachentsev I. D., 2003, Astrophysics, 46, 144 Nilson P., 1974, in UGCA, pp. 0–+ Rizzi L., Tully R. B., Makarov D., Makarova L., Dolphin A. E., Sakai S., Shaya E. J., 2007, ApJ, 661, 815 Ry´sA., Grocholski A. J., van der Marel R. P., Aloisi A., Annibali F., 2011, ArXiv e-prints Salpeter E. E., 1955, ApJ, 121, 161 Schlegel D. J., Finkbeiner D. P., Davis M., 1998, ApJ, 500, 525 Sharina M. E., Karachentsev I. D., Dolphin A. E., Karachentseva V. E., Tully R. B., Karataeva G. M., Makarov D. I., Makarova L. N., Sakai S., Shaya E. J., Nikolaev E. Y., Kuznetsov A. N., 2008, MNRAS, 384, 1544 Stil J. M., Gray A. D., Harnett J. I., 2005, ApJ, 625, 130 Stil J. M., Israel F. P., 1998, A&A, 337, 64 —, 2002, A&A, 392, 473 Tully R. B., Rizzi L., Dolphin A. E., Karachentsev I. D., Karachentseva V. E., Makarov D. I., Makarova L., Sakai S., Shaya E. J., 2006, AJ, 132, 729 Vaduvescu O., McCall M. L., Richer M. G., Fingerhut R. L., 2005, AJ, 130, 1593 Vallenari A., Bomans D. J., 1996, A&A, 313, 713 Walter F., Brinks E., de Blok W. J. G., Bigiel F., Kennicutt Jr. R. C., Thornley M. D., Leroy A., 2008, AJ, 136, 2563

A This paper has been typeset from a TEX/ LTEX file prepared by the author.

c XXX RAS, MNRAS 000, 1–11